Slackbot
10/05/2023, 6:49 PMJosé Morales
10/05/2023, 6:55 PMMatej
10/05/2023, 6:55 PMMatej
10/05/2023, 6:56 PMMatej
10/05/2023, 6:57 PMcv_results = fcst.cross_validation(
df_test,
static_features=[],
n_windows=2,
h=horizon,
step_size=1,
refit=False,
fitted=True
)
but id like to simply go over the entire dataset, somehow I am unable to set up n_windows correctlyJosé Morales
10/05/2023, 6:58 PMstatic_features
argument. By providing an empty list you're saying none are static (they all change over time) which is probably causing a bad join somewhereMatej
10/05/2023, 6:59 PMMatej
10/05/2023, 6:59 PMMatej
10/05/2023, 7:00 PMJosé Morales
10/05/2023, 7:01 PMMatej
10/05/2023, 7:02 PMJosé Morales
10/05/2023, 7:03 PMMatej
10/05/2023, 7:04 PMMatej
10/05/2023, 7:06 PMMatej
10/05/2023, 7:07 PMJosé Morales
10/05/2023, 7:09 PMinput_size
but you have to be careful with the number of samples there, because if you're using lag5 for example and dropna=True
then the features drop 5 rows, so you'll need input_size=6
to get a single sample at the endJosé Morales
10/05/2023, 7:09 PMMatej
10/05/2023, 7:29 PMmax_horizon: int, optional (default=None)
Train this many models, where each model will predict a specific horizon.
• it seems this is for something else. For when I assume different dynamics for each time horizon it seems.Matej
10/05/2023, 7:41 PMbefore_predict_callback
after_predict_callback
José Morales
10/05/2023, 7:46 PMMatej
10/05/2023, 7:50 PM